Automatically Mapping Organizations to Addresses

1. A computer-implemented method comprising: storing a plurality of membership records and a plurality of access records, wherein each membership record of the plurality of membership records comprises an organization identifier that identifies an organization and a member identifier that identifies a member, wherein each access record of the plurality of access records comprises a member identifier that identifies a member and an internet protocol (IP) address that identifies a host computer; identifying a localized subset of access records of the plurality of access records, wherein each access record in the localized subset has a particular IP address or IP address subnet; identifying an involved subset of membership records of the plurality of membership records, wherein each membership record in the involved subset has a member identifier that matches a member identifier of an access record of the localized sub set; determining statistical information based on the localized subset of access records and the involved subset of membership records; and one or more computers identifying, based on the statistical information, a particular organization identifier that occurs in the membership records of the involved subset with a frequency that exceeds a threshold; associating, by the one or more computers, the particular organization identifier with the particular IP address or IP address subnet; indicating that a membership record, of the plurality of membership records, that is not in the involved subset of membership records incorrectly has the particular organization identifier.

2. The method of claim 1 further comprising categorizing the particular IP address as one of: commercial, private, or public.

3. The method of claim 2 wherein categorizing the particular address comprises categorizing the particular address based on the statistical information.

4. The method of claim 3 wherein categorizing the particular address comprises categorizing the particular address based on an entropy of the organization identifiers of the membership records of the involved subset.

5. The method of claim 3 wherein categorizing the particular address comprises categorizing the particular address based on a size of the involved subset of membership records.

6. The method of claim 1 wherein the statistical information comprises an entropy of the organization identifiers of the membership records of the involved sub set.

7. The method of claim 1 wherein the statistical information comprises a percentage, of the membership records of the involved subset, having an organization identifier that matches the organization identifier.

8. The method of claim 1 further comprising determining, based on the statistical information, a confidence level that the organization identifier of a particular membership record of the plurality of membership records is accurate.

9. The method of claim 1 wherein each access record of the plurality of access records further comprises a timestamp, and all access records of the localized subset have a timestamp that is more recent than a particular time.

10. The method of claim 1 wherein the statistical information is determined only if a size of the involved subset of membership records exceeds a threshold.

11. The method of claim 1 wherein the statistical information is determined only if a size of an affiliated subset of membership records of the plurality of membership records exceeds a threshold, and all membership records of the affiliated subset have an organization identifier that matches the particular organization identifier.

12. The method of claim 1 wherein the IP address of each access record of the plurality of access records comprises an IP version 6 (IPv6) address, wherein the IPv6 address of each access record of the plurality of access records comprises a subnet, and wherein all access records of the localized subset that have the particular address consist of all access records of the localized subset that have a particular subnet.

13. The method of claim 1 wherein each access record of the plurality of access records represents a webpage view.

14. The method of claim 1 wherein the particular address belongs to a particular organization, wherein the method further comprises iteratively adjusting at least one threshold until the organization identifier matches an identifier of the particular organization.

15. A system comprising: a storage system configured to store a plurality of membership records and a plurality of access records, wherein each membership record of the plurality of membership records comprises an organization identifier that identifies an organization and a member identifier that identifies a member, wherein each access record of the plurality of access records comprises a member identifier that identifies a member and an internet protocol (IP) address that identifies a host computer; and a processor connected to the storage system and configured to: identify a localized subset of access records of the plurality of access records, wherein each access record in the localized subset has a particular IP address or IP address subnet; identify an involved subset of membership records of the plurality of membership records, wherein each membership record in the involved subset has a member identifier that matches a member identifier of an access record of the localized sub set; determine statistical information based on the localized subset of access records and the involved subset of membership records; and identify, based on the statistical information, a particular organization identifier that occurs in the membership records of the involved subset with a frequency that exceeds a threshold; associate the particular organization identifier with the particular IP address or IP address subnet; indicate that a membership record, of the plurality of membership records, that is not in the involved subset of membership records incorrectly has the particular organization identifier.

16. One or more non-transitory computer readable media comprising instructions that when executed by one or more processors cause: storing a plurality of membership records and a plurality of access records, wherein each membership record of the plurality of membership records comprises an organization identifier that identifies an organization and a member identifier that identifies a member, wherein each access record of the plurality of access records comprises a member identifier that identifies a member and an internet protocol (IP) address that identifies a host computer; identifying a localized subset of access records of the plurality of access records, wherein each access record in the localized subset has a particular IP address or IP address subnet; identifying an involved subset of membership records of the plurality of membership records, wherein each membership record in the involved subset has a member identifier that matches a member identifier of an access record of the localized sub set; determining statistical information based on the localized subset of access records and the involved subset of membership records; identifying, based on the statistical information, a particular organization identifier that occurs in the membership records of the involved subset with a frequency that exceeds a threshold; associating the particular organization identifier with the particular IP address or IP address subnet; indicating that a membership record, of the plurality of membership records, that is not in the involved subset of membership records incorrectly has the particular organization identifier.

17. The one or more non-transitory computer readable media of claim 16 wherein the instructions further cause categorizing, based on the statistical information, the particular address as one of: commercial, private, or public.

18. The one or more non-transitory computer readable media of claim 16 wherein the statistical information comprises an entropy of the organization identifiers of the membership records of the involved subset.

19. The one or more non-transitory computer readable media of claim 16 wherein the statistical information comprises a percentage, of the membership records of the involved subset, having an organization identifier that matches the organization identifier.